TITweetIndex

Cabinet Office @cabinetofficeuk · 26 Aug 2026

How do we keep Britain safe as AI becomes more powerful? @KanishkaNarayan explains how the @AISecurityInst recently tested an AI system that tried to go beyond its instructions, and how it was stopped before any real-world harm was caused. https://t.co/E2aegpX8XW
18 147Views
23Likes
6Reposts
18Replies
4Quotes
6Bookmarks

Is that a lot?

1.00×vs this author's median18 147 views is typical
40Percentile for this authorof 5 recent posts
0.81×vs 100K–1M median22 338 views is typical
4.60%Reachviews ÷ followers
0.28%Engagement rateof viewers reacted

Compare with the benchmark table →

Open on X →